Front-end Analog Pre-Processing for Real Time Psychophysiological Stress Measurements

Every day computers become more influential in our daily lives. In an attempt to improve our interaction with computers, the emerging field of Affective Computing strives to provide the necessary mechanisms that will make machines aware of the affective state of their users. This paper explains the challenges in pre-processing psychophysiological signals, including Blood Volume Pulse (BVP) , Galvanic Skin Response (GSR) and Skin Temperature (ST) for the determination of the emotional state of a computer user, particularly, as he/she experiences emotional stress. These signals are appropriate because they are non-invasive, non-obtrusive and their variation under stress is predictable. The Galvanic Skin Response measures the change in electrodermal activity (increase in conductance) as sweat glands are stimulated to produce a hydrate solution. The Blood Volume Pulse waveform reflects modifications in heart rate, stroke volume and peripheral cardiovascular resistance modulated by the sympathetic nervous system. The acral Skin Temperature experiences short term changes that originate from vasomotor activity in the arterioles. The following sections describe the instrumentation setup that has been developed to monitor these three variables as computer users experience varying levels of stress, elicited by the completion of a series of “Stroop Test”.